500 Dispatch

Genome sequencing: The complete code for a eukaryotic cellMark Johnston

The complete sequencing of the genome of a simpleeukaryotic organism — the budding yeastSaccharomyces cerevisiae — is a milestone for biology,and sets the stage for a complete understanding of howa eukaryotic cell functions.

Address: Department of Genetics Box 8232, Washington UniversityMedical School, 4566 Scott Avenue, St. Louis, Missouri 63110, USA.

Current Biology 1996, Vol 6 No 5:500–503

© Current Biology Ltd ISSN 0960-9822

The belief that research on the budding yeast, Saccha-romyces cerevisiae, would reveal processes fundamental toall eukaryotic cells drove its establishment as one of thepremier ‘model organisms’ of experimental biology. Thatpromise has been fulfilled beyond all expectations overthe last 40 to 50 years. The pioneers in the field were ini-tially attracted to yeast by the ease with which it can begenetically manipulated [1]; the subsequent developmentof ways of manipulating the yeast genome at will [2,3]attracted a further large cadre of investigators in the 1980s.Yeast has become so popular, and work with it so produc-tive, that it was hard for many of us to imagine that itcould become any more attractive as an experimentalorganism. But that is exactly what has happened, with theannouncement on 24 April that the S. cerevisiae genomehas been completely sequenced. This achievement isalready attracting even more scientists to yeast, and itensures that they will be even more productive than theirpredecessors. Furthermore, it encourages us to pursueeven more vigorously the goal that has been implicit fromthe beginning: the complete understanding of how aeukaryotic cell functions. The attainment of this lofty goalnow seems possible.

The yeast genome project — the goal of which was toidentify and map all the genes of the organism — began inearnest in the 1950s [4]. It was not until 1985 that it wasclear that the yeast nucleus contains 16 chromosomes [5],and 769 genes were genetically mapped on them [6]. Theproduction of sets of mapped fragment clones spanningnearly the entire 14–15 megabase (Mb) yeast genome [7,8]set the stage for the ultimate genome mapping — determi-nation of the complete DNA sequence — which has nowbeen accomplished through an international collaborativeeffort of many laboratories. More than half of the sequence(55 %) was completed by a large network of more than 70laboratories in Europe [9–13]. Significant contributionsfrom other groups — including the Sanger Centre inEngland (17 %), Washington University in St. Louis (15 %)

[14], Stanford University (7 %), McGill University in Mon-treal (4 %) [15], and RIKEN University in Japan (2 %) [16]— enabled timely completion of the first eukaryoticgenome sequence. The well-coordinated internationaleffort has proved to be very efficient, with surprisinglylittle duplication of effort. The few instances when asequence was determined by two groups (about 3 % of thetotal) was either deliberate, to test sequence quality, or theresult of unavoidable overlaps between clones. It can onlybe hoped that the success of the yeast community inworking together to achieve an important goal in a timelyand efficient manner, while maintaining high scientificstandards, will be replicated by those tackling largergenomes.

The project has met high standards. The sequence wasconsidered complete only when it had been determinedon both strands, with no gaps. Furthermore, the sequenceof each chromosome was determined all the way out to thetelomeres (the G1–3A repeats at the ends of each chromo-some) [17]. We can thus say that every nucleotide of theyeast genome has been sequenced. We believe that thisdegree of completion is desirable, given the level of inter-est in the yeast sequence, and the consequent closescrutiny of it that is expected. The accuracy of thesequence is also very high. All groups (but especially theEuropean network) have expended a great deal of effort toensure sequence accuracy, and it has paid off. Forexample, a comparison of 172 kilobases (kb) of sequencedetermined independently in Europe and St. Louisrevealed only 14 discrepancies. More than half of thesewere due to unavoidable strain or clone polymorphisms(G. Volkaert, personal communication), so the estimatederror rate is significantly less than one mistake in 10 kb.Other independent estimates of the error rate yieldedslightly higher values [12,13]. (It is worth noting that thisis ~5–10-fold more accurate than the piecemeal contribu-tions of yeast sequence in the public databases).

What have we learned from the complete sequence of theyeast genome? We now know that the yeast nucleargenome consists of ~12.5 Mb of unique DNA thatencodes the ~6000 proteins that constitute this simpleeukaryotic cell (exclusive of the 1–2 Mb rDNA repeatcluster, and a few other repeat families, such as the ~30 kb‘CUP1’ repeats). We expected the genes to be closelypacked, but having the opportunity to view the sequenceof entire chromosomes [10–16] impresses us with the highinformation content of the sequence: about 70 % of itcodes for protein, with a gene approximately every 2 Kb.We know from piecemeal sequencing that only about

Dispatch 501

4.5 % of yeast genes contain introns [18], nearly all ofwhich are small and located at the extreme 5′ end of genes[19], and systematic sequencing has not altered this view.Genes are roughly equally distributed on both DNAstrands, and they appear to be randomly orientated withrespect to each other. The average gene size is ~480codons, about half as large again as an average gene of thebacterium Escherichia coli. The largest protein (L8004.13),is predicted to contain 4911 amino acids, while the short-est, plasma membrane proteolipid 1 (PMP1), is predictedto contain 38 amino acids.

Perhaps the most startling revelation of the completeyeast genome sequence is that less than half of all thegenes have previously been identified through geneticand biochemical analysis, despite intensive analysis overmany years by the large and highly productive yeast com-munity. As of March 23, 1996, only about 44 % of the com-plete gene set had been characterized experimentally —2307 of the 5271 genes tabulated by Jim Garrels and pre-sented in the Yeast Protein Database (http://www.pro-teome.com/YPDhome.html) [20]. Systematic sequencinghas thus yielded a wealth of previously unseen genes, thefunctions of which must be determined if we are to claimto understand how a simple eukaryotic cell functions. Thefunctions of some of the novel gene products can beguessed from their similarities to other proteins: 30 %(899) of the 2964 previously unrecognized genes showsome sequence similarity to other, characterized proteins.However, that leaves 39 % of yeast genes (~2300) forwhich there are no clues as to their function. Whileincreasingly sensitive homology searches may provideinsights into the function of some of these genes [21,22], itis clear that a large fraction of predicted yeast proteins will

not give up their secrets easily. The challenge is to figureout what they do.

One of the things that has probably hindered the geneticidentification of all yeast genes is the fact that some ofthem are redundant. Geneticists have always worriedabout this, but until now have not known how serious aproblem it is. Analysis of the first few chromosomes to besequenced showed that about 5 % of the genes are theresult of recent duplications [10–12,14]. If the criteria forclassifying genes as redundant is relaxed, the apparentlevel of redundancy will of course increase, and moregene pairs will be recognized once the entire sequence isanalyzed. Some people believe as many as 30 % of allgenes are duplicated [23]. Now that we are able to recog-nize redundant genes, we should be able to address thefunctions of some of these previously inaccessible genes.

Even though the analysis of the yeast genome sequence isstill far from complete, it is clear what kinds of proteins,and roughly how many of each type constitute this simpleeukaryotic cell (Table 1). A large number of proteins seemto function in the nucleus: the major class of proteins bysubcellular localization reside in the nucleus; the majorclass by functional category are transcription factors; and alarge class by molecular environment are DNA-associated.This is perhaps not surprising, as the organism mustdevote much of its effort to the replication, maintenance,proper segregation, and regulation of expression of its 6000genes. We are also not surprised that the major class of pro-teins according to molecular environment are integralmembrane proteins, many of which are transporters [24],as a yeast cell must obtain diverse nutrients from, andcommunicate with, its environment. It seems curious,

Table 1

Yeast proteins categorized as of March 23, 1996.

By functional category By major localization category By molecular environment

152 Transcription factors 539 Nuclear proteins 549 Integral membrane proteins109 Protein kinases 420 Cytoplasmic proteins 383 DNA-associated proteins72 Amino-acid metabolism enzymes 274 Mitochondrial proteins 194 Soluble proteins50 GTPases 110 Plasma membrane proteins 166 Ribosomal proteins37 Proteases (non-proteosomal) 77 Endoplasmic reticulum proteins 104 RNA-associated proteins35 Protein phosphatases 51 Cytoskeletal proteins 99 Peripheral membranes34 Heat-shock proteins 54 Extracellular and cell-wall proteins 24 Actin-associated proteins34 tRNA synthetases 51 Vacuolar proteins 16 Tubulin-associated proteins24 Serine-rich protein family 31 Golgi apparatus proteins 44 Protein synthesis factors21 Conserved ATPase domain 23 Peroxisomal proteins20 Cyclins 28 Vesicular, secretory system proteins20 ATP-binding cassette (ABC)16 Glucose metabolism enzymes 139 Unspecified membrane proteins14 Proteasome subunits 3492 Unknown localization12 Ubiquitin-conjugating enzymes11 GTPase-activating proteins8 Guanine nucleotide exchange factors

This data is updated daily: see <http://www.proteome.com/YPD_contents_by_category.html.>

502 Current Biology 1996, Vol 6 No 5

however, that an organism that can live without carryingout oxidative phosphorylation has a large class of proteinsthat are localized in the mitochondria.

What should we do with the wealth of data the yeastgenome sequence provides? We should, of course,examine it, and fortunately there are several excellentdatabases that allow us to do so easily (Table 2). Thesedatabases, which for the most part are complementary, aresure to facilitate efficient and productive use of thesequence information. The databases are currently beingproductively used by the community of yeast scientists,but their major impact is likely to be in the guidance theywill provide to scientists working on other organisms,including Homo sapiens, who come across yeast proteinssimilar to their proteins of interest.

We need to begin to try to find out what the products ofeach of the ~3300 new genes uncovered by systematicsequencing do. It is not entirely clear how to go about thistask. Some kind of systematic analysis will probably benecessary to begin to understand the roles of genes thatprovide no hint of their function [25]. Such an approachhas been recently organized by the Europeans, as the nextphase of their successful sequencing network, and otherad hoc efforts are underway [26,27]. Many genes will bedisrupted, and a standardized battery of tests carried outto determine the effect of gene loss on the organism. Thisshould yield a wealth of information that is expected toilluminate the functions of some of these genes.

It is likely, however, that this enterprise will encounter asignificant problem, as it is expected that disruption of alarge fraction of yeast genes (about 50 %) will not haveany obvious effect on the organism, making it difficult tothink of further experiments that might reveal their func-tion. Yeast researchers will, nevertheless, undoubtedlyalight upon some of these genes in the course of their

ongoing studies, using the variety of powerful techniquesavailable to them — the two-hybrid method for detectingprotein–protein interactions, for example. As the com-plete sequence of the gene will be on hand, the researchercan proceed immediately with further incisive experimen-tation, suggested by his or her avenue of the research,which may reveal the function of the protein encoded bythe gene. We also expect that the function of many ofthese genes will be illuminated by analysis of their homo-logues in other ‘model’ organisms, most of which arericher in phenotypes than yeast, which may provide valu-able clues to functions of the encoded proteins [28,29].

A complete understanding of how a simple eukaryotic cellfunctions will require information from many sources. Wehope that the recent completion of the genome sequencewill bolster the efforts of the yeast community and attractnew investigators to our fold, because our next task —understanding how a simple eukaryotic cell functions — ismuch more difficult than the task we have just completed.We look forward to the challenge.

AcknowledgementsI thank André Goffeau for comments on the manuscript, and for having theforesight to initiate the sequencing project, the skill to organize thesequencers, the determination to maintain high standards, and the fortitudeto lead us to the finish line. All of us in the yeast community owe him a greatdebt. I also thank Jim Garrels for maintaining an excellent database andallowing me to include the information here.

References1. Hall MN, Linder P: The Early Days of Yeast Genetics. Cold Spring

Harbor, New York: Cold Spring Harbor Press; 1993. 2. Hinnen A, Hicks JB, Fink GR: Transformation of yeast. Proc Natl

Acad Sci USA 1978, 75:1929–1933.3. Scherer S, Davis RW: Replacement of chromosome segments

with altered DNA sequences constructed in vitro. Proc Natl AcadSci USA 1979, 76:4951–4955.

4. Hawthorne DC, Mortimer RK: Chromosome mapping inSaccharomyces: centromere-linked genes. Genetics 1960,45:1085–1110.

5. Carle GF, Olson V: An electrophoretic karyotype for yeast. ProcNatl Acad Sci USA 1985, 82:3756–3760.

6. Mortimer RK, Schild D: Genetic map of Saccharomyces cerevisiae,Edition 10. Yeast 1989, 5:321–403.

7. Riles L, Dutchik JE, Baktha A, McCauley BK, Thayer EC, Ledkie MP,Braden VV, Depke JE, Olson MV: Physical maps of the six smallestchromosomes of Saccharomyces cerevisiae at a resolution of 2.6kilobase pairs. Genetics 1993, 134:81–150.

8. Thierry A, Gaillon L, Galibert R, Dujon B: Construction of a completegenomic library of Saccharomyces cerevisiae and physical mappingof chromosome XI at 3.7 kb resolution. Yeast 1995, 11:121–135.

9. Goffeau A, Vassaroti A: The European project for sequencing theyeast genome. Res Microbiol 1991, 142:901–903.

10. Oliver SG, van der Aart QJ, Agostoni-Carbone ML, Aigle M,Alberghina L, Alexandraki D, Antoine G, Anwar R, Ballesta JP, Benit P,et al.: The complete DNA sequence of yeast chromosome III.Nature 1992, 357:38–46.

11. Dujon B, Alexandraki D, Andre B, Ansorge W, Baladron V, Ballesta JP,Banrevi A, Bolle PA, Bolotin-Fukuhara M, Bossier P, et al.: CompleteDNA sequence of yeast chromosome XI. Nature 1994,369:371–378.

12. Feldmann H, Aigle M, Aljinovic G, Andre B, Baclet MC, Barthe C,Baur A, Becam AM, Biteau N, Boles E, et al.: Complete DNAsequence of yeast chromosome II. EMBO J 1994, 13:5795–5809.

13. Galibert F, Alexandraki A, Baur A, Boles E, Chalwatzis N, Chuat J-C,Coster F, Cziepluch C, De Haan M, Domdey H, et al.: Completenucleotide sequence of Saccharomyces cerevisiae chromosomeX. EMBO J 1996, 15:2031–2049.

Table 2

Yeast genome databases.

Database name URL

SGD: Saccharomyces Genome http://genome-www.stanford.eduDatabase

YPD: Yeast Protein Database http://www.proteome.com

MIPS: European network http://speedy.mips.biochem.informatics mpg.de/mips/yeast/

XREFdb: Cross-referencing yeast http://www.ncbi.nlm.nih.gov/and other genomes XREFdb/

WWW Virtual Library: Yeast http://genome-www.stanford.edu/VL-yeast.html

GeneQuiz: Comprehensive http://genecrunch.sgi.com/homology search results

14. Johnston M, Andrews S, Brinkman R, Cooper J, Ding H, Dover J, DuZ, Favello A, Fulton L, Gattung S, et al.: Complete nucleotidesequence of Saccharomyces cerevisiae chromosome VIII. Science1994, 265:2077–2082.

15. Bussey H, Kaback DB, Zhong W, Vo DT, Clark MW, Fortin N, Hall J,Ouellette BF, Keng T, Barton AB, et al.: The nucleotide sequence ofchromosome I from Saccharomyces cerevisiae. Proc Natl Acad SciUSA 1995, 92:3809–3813.

16. Murakami Y, Naitou M, Hagiwara H, Shibata T, Ozawa M, SasanumaS, Sasanuma M, Tsuchiya Y, Soeda E, Yokoyama K, et al.: Analysis ofthe nucleotide sequence of chromosome VI of Saccharomycescerevisiae. Nat Genet 1995, 10:261–268.

17. Louis E, Borts R: A complete set of marked telomeres inSaccharomyces cerevisiae for physical mapping and cloning.Genetics 1995, 139:125–136.

18. Rodriguez-Medina JR, Rymond BC: Prevalence and distribution ofintrons in non-ribosomal protein genes of yeast. Mol Gen Genet1994, 243:532–539.

19. Fink GR: Pseudogenes in yeast? Cell 1987, 49:5–6.20. Garrels JI: YPD — A database for the proteins of Saccharomyces

cerevisiae. Nucleic Acids Res 1996, 24:46–49. 21. Koonin EV, Bork P, Sander C: Yeast chromosome III: new gene

functions. EMBO J 1994, 13:493–503.22. Ouzounis C, Bork P, Casari G, Sander C: New protein functions in

yeast chromosome VIII. Prot Sci 1995, 4:2424–2428.23. Richard G-F, Fairhead C, Dujon B: Complete transcriptional map of

yeast chromosome XI in different life conditions. J Mol Biol, inpress.

24. Nelissen B, Mordant P, Jonniaux JL, De Wachter R, Goffeau A:Phylogenetic classification of the major superfamily of membranetransport facilitators, as deduced from yeast genome sequencing.FEBS Lett 1995, 377:232–236.

25. Oliver SG: From DNA sequence to biological function. Nature1996, 379:597–600.

26. Burns N, Grimwade B, Ross-Macdonale PB, Choi EY, Finberg K,Roeder GS, Snyder M: Large-scale analysis of gene expression,protein localization, and gene disruption in Saccharomycescerevisiae. Genes Dev 1994, 8:1087–1105.

27. Smith V, Botstein D, Brown PO: Fenetic footprinting: a genomicstrategy for determining a gene’s function given its sequence.Proc Natl Acad Sci USA 1995, 92:6479–6483.

28. Tugenreich S, Boguski MS, Seldin MS, Hieter P: Linking yeastgenetics to mammalian genomes: identification and mapping ofthe human homolog of CDC27 via the expressed sequence tag(EST) data base. Proc Natl Acad Sci USA 1993, 90:10031–10035.

29. Tugenreich S, Bassett Jr. DE, McKusick VA, Boguski MS, Hieter P:Genes conserved in yeast and humans. Hum Mol Genet 1994,3:1509–1517.

Dispatch 503